INDEX

Explanations

phrases related to cognition and mental states

oai_token-act-pair · gpt-3.5-turbo

words related to cognition and diagnostic terms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

netflix

-0.77

uates

-0.75

fters

-0.72

ting

-0.71

ulously

-0.70

ted

-0.70

FORE

-0.67

pee

-0.67

keyes

-0.67

eed

-0.67

POSITIVE LOGITS

ostic

0.93

ostics

0.93

 Skydragon

0.80

 guiActiveUnfocused

0.75

acent

0.75

obil

0.74

auts

0.69

atorium

0.68

ancy

0.68

atism

0.68

Activations Density 0.052%

phrases related to cognition and mental states

words related to cognition and diagnostic terms

No Comments

No Known Activations